The LiveLabs Testbed & Mobile Sensing-based Applications. Archan Misra, Singapore Management University

advertisement
The LiveLabs Testbed & Mobile Sensing-based
Applications.
Archan Misra,
Singapore Management University
archanm@smu.edu.sg
Feb 17, 2012
UMD, 2012
Talk Outline
• LiveLabs: A Mobile Behavioral Experimentation Analogue
of PlanetLab
• Energy-Efficient Context Acquisition
• A3R: Adaptive Accelerometer-based Activity Recognition
• ACQUA and Distributed Analytics
• Using Rich, Individual Context
• Context-Driven Real-time Femtocell Adaptation
• CAMEO: Predicting Context for Better Mobile Advertising
UMD, 2012
LiveLabs
Globally-unique lifestyle R&D
1. Network technologies for
advanced broadband wireless
infrastructure.
2. An automated service that lets
consumer companies easily run
lifestyle experiments.
3. A participant base of 30,000
consumers in 3 key public space
(SMU, Malls, Sentosa)
Globally-First Automated
Behavioral
Experimentation Service
UMD, Feb 2012
LiveLabs
Globally-Unique Individual
and Aggregate UsageAdaptive Wireless Network
3
LiveLabs – Downtown Lifestyle Sensing
Tourism & Hospitality
LiveLabs@Sentosa
• Crowd Behavior &
Movement Optimization
• Personalized
Recommendations for
Leisure and F&B
Downtown Lifestyle Sensing Testbed:
•
•
LiveLabs@
SMU
Wireless infrastructure that adapts to
real-time usage & hotspots
Behavioral experimentation software
LiveLabs@ LiveLabs@
Clarke Quay Plaza Sing
LiveLabs Ecosystem
Examples of
Expected
Future Users
Committed
Users
Globally-First Automated
Behavioral
Experimentation Service
Key Technology
Providers & Users
UMD, Feb 2012
LiveLabs
Globally-Unique
Individual and Aggregate
Usage-Adaptive Wireless
Network
Key R&D Challenges and Outcomes
• Challenge 1: Deep, continuous, context collection
– Year 1: Collect context from network traces only
– Year 2: Collect some context from cell phones
– Year 3: Energy-efficient deep context (cell phones + network)
• Challenge 2: Fine grained indoor localization
– Year 1: 5 to 10m resolution
– Year 2: 2 to 5m resolution
– Year 3: <= 1m resolution
• Challenge 3: Handle transient network traffic loads
– Year 1: Offload pre-determined network loads to wired backbone
– Year 2: Offload network loads to wireless backbones
– Year 3: Offload traffic based on dynamic traffic patterns
• Challenge 4: Run automated social experiments on cell phones
– Year 1: Build basic framework to run experiments
– Year 2: Integrate mechanisms to control participant selection
– Year 3: Integrate end-to-end tools to allow 3rd party developers to use
LiveLabs experimentation service
• Challenge 5: Support privacy preferences of users at runtime
– Year 2: Build in mobile device support for privacy enforcement
– Year 3: Dynamic App checking to enforce context-sensitive privacy
UMD, Feb 2012
6
EXTERNAL (3RD PARTY)
APPLICATIONS
BEHAVIORAL EXPERIMENTS
LiveLabs Architecture
Behavioral Experimentation
Front-End
Participant
Selection
Service
SERVICE DELIVERY PLATFORM
App
Validation
Service
MME/S-GW
Participant
Preference DB
App
Deployment
Service
GGSN
RG-SGSN
AAA-SERVER
COMMERCIAL PROVIDER
NETWORK CORE
PROVISIONING
SUB-SYSTEM
Experiment
Regulation
Service
App Runtime
Monitoring
Service
Cloudlet Server
(Realtime Edge Analytics)
Cloudlet Server
(Realtime Edge Analytics)
Femto Gateway
Controller
MacroCell-Controller
Context API
RUNTIME
SUB-SYSTEM
UMD, Feb 2012
Participant
Context
Repository
Macro BS
Femto BS
7
Talk Outline
• LiveLabs: A Mobile Behavioral Experimentation Analogue
of PlanetLab
• Energy-Efficient Context Acquisition
• A3R: Adaptive Accelerometer-based Activity Recognition
• ACQUA and Distributed Analytics
• Using Rich, Individual Context
• Context-Driven Real-time Femtocell Adaptation
• CAMEO: Predicting Context for Better Mobile Advertising
UMD, 2012
A3R: Adaptive Accelerometer-based Activity
Recognition
• Key Idea: Adjust accelerometer ―parameters‖
based on the current activity of the individual.
• Two parameters:
– Sampling frequency of accelerometer stream (sf)
– Features Used for Activity Classification (F)
• Goal: reduce energy overhead of activity
recognition without sacrificing accuracy
UMD, Feb 2012
9
Energy Overhead Variation
50
45
Energy (Joules)
40
35
30
25
T-Domain + F-Domain
20
Only T-Domain
15
10
5
0
0
5
10
15
20
25
30
Frequency (Hz)
• Energy overhead increases with sf.
• Non-linear increase when frequency-domain features are
selected along with time-domain features.
UMD, Feb 2012
10
Classification Accuracy Variation
Time-Domain Only
• Most ‗stationary‘ activities (e.g., sit, stand) OK with only sf
(1/0.5 Hz).
• Selected activities (e.g., climbing stairs) require
(time,frequency) features
UMD, Feb 2012
11
A3R: Results on Real User Behavior
• Over
30% savings in energy
under “regular” lifestyle
UMD, Feb 2012
12
ACQUA (Acquisition Cost-Aware Query
Adaptation) Scenario
Context deduced from
wirelessly connected
sensors+ sensors on tother
phones
Phone runs a complex
event processing (CEP)
engine with rules for
alerts
SPO2
ECG
HR
Temp.
Acc.
...
IF Avg(Window(HR)) > 100
AND Avg(Window(Acc)) < 2
AND AVG(Window(Temp))>80F
THEN SMS(caregiver)
13
ACQUA Architecture
Cost Modeler
Query Logic Specification
Module
• Stream-SQL based specification
• External specification of sensorspecific trx. Cost model
• Dynamic evaluation of stream
selectivity
of query syntax
Normalized
Query Syntax
C(.); P(.)
Dynamic Query
Evaluation Optimizer
• Signals sensors to adjust pushvs-pull mode
• Determines retrieval sequence
for sensor streams
Asynchronous Event
Engine
• Maintains partial
query evaluation state
Push/Pull, Batch commands
Dynamic Sensor
Control (DSC)
14
Acquiring N Data-Tuples from Sensor
Power
Pa
Pi
Idle
S
w
i
t
c
h
Active
Time
N*S/B
• Idle mode
consumes Pi mW
• Active mode
consumes Pa mW
• Sensor rate is f Hz
• A tuple is S bits
• Bandwidth is B
Mbps
N/f
UMD, Feb 2012
15
6/2/
Enhanced Evaluation Order
if Avg(S2, 5)>20 AND S1<10 AND Max(S3,10)<4 then email(doctor).
Predicate
Avg(S2, 5)>20
S1<10
Max(S3,10)<4
Acquisition
5 * .02 = 0.1 nJ 0.2 nJ
10 * .01 = 0.1 nJ
Pr(false)
0.95
0.5
0.8
Acq./Pr(f)
0.1/0.95
0.2/0.5
0.1/0.8
• Evaluate predicates with lowest energy consumption
first
• Evaluate predicates with highest false probability first
• Evaluate predicate with lowest normalized acquisition
cost first.
UMD, Feb 2012
17
6/2/
Performance Results
802.11
Bytes
Energy
Bluetooth
UMD, Feb 2012
20
6/2/
ProxSense: Distributed Evaluation of CCG Graphs
if Avg(C, 10)<50 AND (AVG(D,5)>3
AND Max(B,4)>100 then
transmit(LocomotionState)
if Avg(A, 5)<70 AND (C<3
OR Max(B,4)>100 then
transmit(location).
AND
Phone 1
AVG(C,10)<50
P1
Phone 2
AND
P1
P2
MAX(B,4)>100
P3
AVG(D,5>3
P2
P3
AND
AVG(C,10)<50
AND
P1
P1
P2
P3
AVG(D,5>3
P2
Remote binding and networked commn.
UMD, Feb 2012
P3
22
6/2/
Talk Outline
• LiveLabs: A Mobile Behavioral Experimentation Analogue
of PlanetLab
• Energy-Efficient Context Acquisition
• A3R: Adaptive Accelerometer-based Activity Recognition
• ACQUA and Distributed Analytics
• Using Rich, Individual Context
• Context-Driven Real-time Femtocell Adaptation
• CAMEO: Predicting Context for Better Mobile Advertising
UMD, 2012
The Femto Problem
•
Handoff when (RSSI(target)RSSI(serving)> Th for a period of Ts)
•
Fixed Th & Ts can mean:
–
–
•
High Th: Fast moving indoor users can
take too long to handoff, leading to loss of
signal quality and throughput at the cell
edge.
Low Th: Slow moving indoor users will
handoff too soon—random movement
can lead to significantly greater ping-pong
effect, especially when signal strength
diffusion is not uniform.
Lot of work in simulations, but very little
captures the practical challenges:
–
–
–
Time-varying, anisotropic, RF
propagation.
Mobile device-based user speed
estimation is not perfect.
No use of prediction of movement
patterns.
UMD, Feb 2012
24
Adaptive High-Bandwidth Indoor Wireless Networks
• Research Questions:
– How to use real-time analytics on
collected context to improve future
wireless network ability to handle
traffic loads?
• Technical novelty:
– Combine network (RF) context +
mobile-device user (RF+sensors+
applications) context to predict
network conditions.
– Dynamically use such current+
predictive group context to adapt
network parameters
UMD, Feb 2012
The Real-Time Closed-Loop Context Sensing &
Adaptation Framework
25
Adaptive Wireless Networks…Progress So Far
• Deployment:
– 6 Femtocell APs deployed on 2
Floors of SIS Building (level 5 and
level 3)
RF Map of
Femto
Deployment
• Emprical Data Collection
– Network conditions and parameters
collected longitudinally
• Research Insights:
– User movement speed strongly
influences network behavior (e.g.,
handoffs)
– Indoor environments require different
analytics than outdoors.
– Two new features provide good
prediction:
• No of ―DL Power Up‖ Signals &
BLER
Active Set Update Helps Predict Handoffs
(Outdoors)
CAMEO: Optimizing Mobile Advertising
Motivation:
• Ad supported free Apps are very
popular.
• Telco
providers
increasingly
moving to metered data plans.
―Free‖ is not really free!
Key Research Idea:
Prefetch Ad Content during Cheap Connectivity (e.g.,
27
WiFi@Home) and serve from local cache on phone
Context & Mechanics of Mobile Advertising
I want
an ad
Typical Context Fields in Ad
Network SDK
Go to
Server
X
Ad ready. Go to Server Y
Load Balanced
Ad Platform
Manager
Send Context Information
Context Sensitive
Ad Provider
Image and ad
host server
Ad
Please Ad
…
sent
Mobile Application
28
CAMEO: Approach & Architecture
• CAMEO ‗predicts‘ user context (location, application use,
etc.)
• CAMEO pre-fetches and caches ads locally using predicted
context, when connected via ―cheap‖ networks.
• Ads served locally when application is invoked.
UMD, Feb 2012
29
Conclusions
• LiveLabs is a large-scale testbed for
– R&D into adaptive context-driven wireless networks and mobile applications
– Easy experimentation with new services over real users in real indoor/outdoor
public spaces.
• Key technical challenges/advances include:
– Energy-efficient sensing and collaboration fusion of activity context
– Using such context to build usage-adaptive heterogeneous access networks
(Macro, femto, WiFi)
– Using such context to optimize application and content delivery architectures.
• Other ongoing projects (not covered here):
– Accurate (<1m) indoor localization without fingerprinting.
– Recognition of semantic activities based on low-level sensor-based signatures.
UMD, Feb 2012
30
Three Key Trends in Mobile Computing
Increased sensorrichness and display
capabilities in mobile
devices
Emergence of
proximal P2P
Among Mobile
Devices
Processing at the
Edge of the Network
(Better responsiveness
& scaling)
UMD, Feb 2012
•Tablet sales to eclipse
laptop sales by 2012.
• Embedded sensor
market doubling each yer
2012-2014
•FlashLinQ radios
from Qualcomm
•SocialWiFi proposal
from WiFi Forum
•Linux-based
processors on gen-2
femtos
•VM-based Cloudlet for
personalized offloading
Source: a) M. Scott Corson, et al,
Towards Proximity-Aware
Internetworking, IEEE Wireless
Commn. Magazine, Dec 2010..
Source: M. Satyanarayana, et al, The
Case for VM-based Cloudlets in
Mobile Computing, IEEE Pervasive
Computing, 2009.
31
Acknowledgements & References
Project
Name
Collaborators
Publication Reference
ACQUA
Lipyeow Lim (Univ. of Hawaii)
A. Misra and L. Lim, ―Optimizing
Sensor Data Acquisition for
Energy-Efficient Smartphonebased Continuous Event
Processing‖, IEEE MDM, 06/2011
A3R
Zhixian Yan, Dipanjan
Chakraborty and Karl Aberer
(EPFL), Vigneshwaran
Subbaraju (SMU)
Under submission.
Femtocell
Adaptation
Srini Seshan (CMU),
Vigneshwaran Subbaraju
(SMU)
---In Preparation.
CAMEO
Srini Seshan (CMU) Azeem
Khan (SMU), Vigneshwaran
Subbaraju (SMU)
HotMobile 2012
UMD, Feb 2012
32
Collaborative Sensing & the ―Mobile
3.0‖ Computing Architecture
Participatory
Sensing
Collaborative
Sensing
•Collaborative for individual
benefit
•Near-real-time analytics
•Localized in space & time
UMD, Feb 2012
33
Download